Automation for Newbs (Myself) Christie Series 2 / Doremi+AP20 / GDC+750

[Harold Hallikainen]

RS232 lines normally idle at -12V or so. Data is sent in one direction on pin 2, and the other on pin 3. If both sides are putting voltage on one pin and not the other, the null modem cable is required. That's a quick check.

[Rick Raskin]

From my RS232 days a millennium ago, crossover 2-3 and 3-2, 20 - 8 and 8-20. That's TX to RX and vice-versa and similarly DTR to DSR. If you are running sync 15 and 17 also apply.

[Ryan Gallagher]

Thanks.
I did find a AP20 manual reference to the CR escape sequence \0d (at least when entering serial strings it should listen for).
Since I tried that sequence in the sending field, I'm leaning towards the cable being the last hurdle.

Will try serial console from a laptop direct to blu-ray tomorrow as well, confirm I can get it responding before adding the AP20/Doremi back into the mix.

Can't win them all in one day.

[Ryan Gallagher]

Null Modem cable was the trick. Have a compliment of blu-ray cues now via the TCP->AP20->Macros->Serial Send route.

Single commands seem to be pretty rock solid. But a chain of them in a macro seems to be problematic even with what seem like excessive delays.
I had a cute "Cue to 0:00 and Pause" macro... but it drops a command every so often and will not end up in the intended state, even with like 2500ms delay between serial sends.

I made sure to disable the Oppo's verbose responses. So unless the handshaking style of this cable is causing issues....
(I have it working with a Null Modem Loopback style as well as a Null Modem Crossover style, but both exhibit same slight issues on strings of commands).

Blu-Ray was totally a bonus for my automation scheme. That serial port might be needed for something else anyway down the line anyhoo.

[Frank Cox]

Would it be worthwhile resending each command to make sure it gets through? Repeat each one say, three times with a short pause in between.

Or just give lots and lots of time between each command. Some commands may take longer to execute than others and if the equipment is "busy" it may not respond.

[Carsten Kurz]

As to the audio fades - the AP20 has a global option for fades between format changes. You can use that option. But: The AP20 does not only support single automation events, but sequences of events, so, you can easily create a macro that implements the fade-out locally on the AP20 upon a special macro call from the Doremi.

[Ryan Gallagher]

As to the audio fades - the AP20 has a global option for fades between format changes. You can use that option. But: The AP20 does not only support single automation events, but sequences of events, so, you can easily create a macro that implements the fade-out locally on the AP20 upon a special macro call from the Doremi.

Carsten, That is essentially what I did in my instinctive approach. The fade sequence is entirely an AP20 macro. Doremi just calls the macro and is free to do other non-AP20 things while it's executing. Format fades may be something to add anyway, I'd just be afraid of a situation where you are changing from one format to another that intentionally has audio immediately. I'm guessing the first part of the new format's audio might get stomped on. Easy to test though.

Frank. It was a bit tricky. The GOTO command puts you on a screen that is expecting certain remote button presses, and exact numbers of them, then a ENTER, and then you have to hide that info display. I tried inserting extra #NU0\0d for additional 0s in the timecode. But when it took the extras it confused it. I may just reinvent that macro with an attempt to target chapter 1 / scene 1 instead of 00:00:00. Less key presses to throw at it. I don't really "need" that functionality in automation, but it was a fun experiment.

I suppose I could also attempt to monitor the AP20's serial output on a terminal. Hard to tell if the hiccup is the AP20 dropping something or the Oppo not taking it. Reduce all my delay times and if all the commands appear it's probably the Oppo's response time that is the issue.

[Ryan Gallagher]

And FYI, i'm only battling with Oppo serial commands because according to docs, their TCP/IP method only allows one client at a time, and I prefer to keep the functionality of the iPad app available too.
(That and it's TCP/IP method requires a discovery command and non-predetermined UDP port after discovery).

[Marco Giustini]

I've used the serial port of the AP20 extensively and it seemed reliable to me. I am not familiar with the Oppo though.

[Ryan Gallagher]

I'll check my serial settings again and make sure AP20 is matching what the Oppo expects and see if I can't get it more reliable.

But onto other "phase maybe" goals, initial research:

Lighting Research:
So we have a Unison Paradigm P-APC Processor for house lighting (on an air-gapped network lighting network). In my initial research only 4 presets can be exposed via Contact Inputs on the processor. It does have a Serial command structure that is available (but that is an extremely long serial run). I'm told, after a configuration tweak, it can accept UDP commands that are the same as the serial structure. We at least have a dry line we could test that with, but permanent solution would likely involve pulling another CAT5. (Or a device like a JNIOR could talk to it's serial, but that still involves pulling another CAT5).

The booth has a 10 Button panel near a 35mm window, the catch is that those 10 buttons are programmed the same as the stage panels, and lack a couple useful film presets. Namely "Curtain Warmers Only" and "Curtain Warmers + Intro Speaker Lights" If I want Warmers my only option is House Full + Warmers. If I want Speaker Lights, I only get House Half + Speaker. If I can retire a few unneeded presets off the booth panel, I could add those two additional ones (Paradigm reprogramming). But adding the automation would be the real boon cause the controller is on the opposite side of the booth etc.

So it all seems doable with some help from the lighting dept, reconfig, and testing. I don't need the entire lighting network accessible, so my instinct would be to have the second NIC on the Doremi be able to talk to it. That or I could maybe change the subnet mask on the AP20 to make the other device reachable and link the switches provided no IP spaces conflict, but in my memory of the AP20 automation options, I don't recall seeing UDP choices. Doremi did have UDP commands for sure.

Main Curtain Control Thoughts:
Here we have a Strong/MDI install with 3 button analog momentary closure control panel in the booth (also near a 35mm window) STOP, OPEN, CLOSE. My novice "path of least resistance" instincts here are to branch off 3 circuits in parallel from the existing booth controller, and utilize some kind of relay or controllable switch via the AP20 TCP or GPIO outputs to alternatively close those momentary circuits.

Relays of course can be very reliable, but my novice understanding is that GPIO voltage is not enough to energize coils on most relays, nor is pulling a pin LOW really what a relay expects for input voltage, so some other device is required for GPIO, or would be more directly TCP controllable and maybe include relays itself. I suppose this is where the JNIORS and other industry gear is typically deployed.

Short of buying a JNIOR, I see lots of other hardware routes to getting these three circuits of control, some would also add TCP/IP methods or the relays themselves baked in too. The price ranges are extremely diverse, some would add up to a JNIOR+industrial relays... but a proof of concept could be done with like 20$ bare smart relay board, since would not be switching a load, just a control circuit.

I think I know what the long term "should do" answer is here. Getting management to agree any of it is necessary is the hurdle. But getting the functionality without the support/warranty totally appears doable on a shoe-string provided we support ourselves and had mgmt's blessing to experiment with curtain automation. None of it is mission critical as long as no relays are capable of failing into a closed state! We still have an operator in the booth that can slap a physical button.

Additional words of wisdom appreciated!

[Mark Gulbrandsen]

I actually second what Steve says about the Eprad eCNA series of automation. I pretty much used them exclusively on almost all installations. They are easy to set up, and or make changes on. Plus not one ever failed for me... EVER! They are also pretty reasonably priced. Both the eCNA 5 and eCNA 10 are good choices. The eCNA 10 would be the choice to interface with more complex lighting systems. The manual for the 5 is in the Film-Tech Warehouse.

[Paul Finn]

Caution: The OPEN and CLOSE curtain motor circuits can be paralleled to multiple locations. The STOP curtain motor circuit CANNOT be paralleled! The STOP circuit is a separate, SERIES circuit that must loop from the motor controller through all operating locations with normally closed (NC) STOP contacts at each control point or automation switching curtain control function, and back to the controller. An interruption any place in the STOP circuit will cause the operating curtain motor to STOP, regardless of where the OPEN/CLOSE command originated. This is also the system's safety feature. The STOP series circuit can also have a "disable" switch in it which can prevent all curtain motor operation. This switch is usually found on the curtain motor controller direct operating panel.

Paul Finn

[Ryan Gallagher]

Caution: The OPEN and CLOSE curtain motor circuits can be paralleled to multiple locations. The STOP curtain motor circuit CANNOT be paralleled! The STOP circuit is a separate, SERIES circuit

Excellent tip there Paul. I found the wiring diagram for our button panels too which confirms stop is in series. But it does demonstrate chaining panels also. So branching off the existing button panel in the booth seems doable. Our two curtain positions created by physical limit switches on the curtain track, so our panels are agnostic in that a open/close/stop can come from any button panel.

Hell even just adding another button panel on the other side of the booth near the digital operator's station would be a good/cheap place to start, and serve to demonstrate the wiring requirements to chain off existing panels.

Our curtain motor system is an ADC 2961/2963 equivalent (suspended setup) with the RCS-1 (Maybe RCS-2) button panels.

While the stop IS indeed wired in series, i'm not 100% clear from the docs if that is a normally open or normally closed circuit. I guess being in series, it would have to be a normally closed... otherwise you'd have to press all the stop buttons simultaneously to change it's state!

Control voltage is 24V (AC I believe according to their catalog, but meter would verify).

So it seems like integration via the existing button panel would require a combination of normally open and normally closed IP or serial controllable relays rated for at least 24vAC (regardless of product used).

Cheers!

[Paul Finn]

I have worked with older ADC 24volt control systems. I neglected to mention that limit switches are also part of the STOP circuit loop, and are necessary to prevent damage to the physical system by stopping the curtain movement at the ends of its open or closed limits of travel.

Paul Finn

[Ryan Gallagher]

I have worked with older ADC 24volt control systems. I neglected to mention that limit switches are also part of the STOP circuit loop, and are necessary to prevent damage to the physical system by stopping the curtain movement at the ends of its open or closed limits of travel.

Paul Finn

Thanks Paul. That makes sense. Effectively tripping the limit is the same as pressing the panel stop button, interrupts the same circuit.

We installed this system ourselves last year when our old curtain finally suffered dry rot and got damaged during a load-in of a taping, but most of the button panels we inherited from the prior ADC setup. Well aware of how dangerous it is to run past limits and hit the hard stops. We won't be "experimenting" again per say unless i'm 100% sure and we have someone standing by the power interrupt, or better yet disengage the drive cog until we have everything 100% working and limits interrupting as intended.